Biodegradation of Chlorobenzene, 1,2-Dichlorobenzene, and 1,4

Mar 8, 2013 - ... of a site contaminated with chlorobenzene (CB), 1,2-dichlorobenzene ... below the first sampling port to create an oxic/anoxic inter...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/est

Biodegradation of Chlorobenzene, 1,2-Dichlorobenzene, and 1,4Dichlorobenzene in the Vadose Zone Zohre Kurt and Jim C. Spain* School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0512, United States S Supporting Information *

ABSTRACT: Much of the microbial activity in nature takes place at interfaces, which are often associated with redox discontinuities. One example is the oxic/ anoxic interface where polluted groundwater interacts with the overlying vadose zone. We tested whether microbes in the vadose zone can use synthetic chemicals as electron donors and thus protect the overlying air and buildings from groundwater pollutants. Samples from the vadose zone of a site contaminated with chlorobenzene (CB), 1,2-dichlorobenzene (12DCB), and 1,4-dichlorobenzene (14DCB) were packed in a multiport column to simulate the interface of the vadose zone with an underlying groundwater plume. A mixture of CB, 12DCB, and 14DCB in anoxic water was pumped continuously through the bottom of column to an outlet below the first sampling port to create an oxic/anoxic interface and a capillary fringe. Removal to below the detection limits by rapid biodegradation with rates of 21 ± 1 mg of CB • m−2 • d−1, 3.7 ± 0.5 mg of 12DCB • m−2 • d−1, and 7.4 ± 0.7 mg of 1.4 DCB • m−2 • d−1 indicated that natural attenuation in the capillary fringe can prevent the migration of CB, 12DCB, and 14DCB vapors. Enumeration of bacteria capable of degrading chlorobenzenes suggested that most of the biodegradation takes place within the first 10 cm above the saturated zone. Biodegradation also increased the upward flux of contaminants and thus enhanced their elimination from the underlying water. The results revealed a substantial biodegradation capacity for chlorinated aromatic compounds at the oxic/anoxic interface and illustrate the role of microbes in creating steep redox gradients.



fringes of plumes.10,14,15 In contrast, biodegradation of synthetic contaminants at the capillary fringe is poorly understood. Biodegradation of volatile petroleum hydrocarbons in the unsaturated zone has been studied extensively by measuring hydrocarbon and oxygen concentrations.8,16 Models were developed for the evaluation of biodegradation based on firstorder decay8,17 to assess hydrocarbon vapor intrusion and predict indoor air contamination.18,19 The results indicated a rapid and predictable biodegradation of hydrocarbons in the presence of oxygen, and therefore oxygen limited biodegradation was incorporated in the vapor intrusion models.19,20 Recently Ma et al. used the Devaull model to show that methane biodegradation was dependent on oxygen concentrations in the vadose zone and that activity was high in the capillary zone in a laboratory system.21 The models and their applications provided an effective tool to predict whether the biodegradation was sufficient to protect the air above the surface, but much of the work to date has focused on light nonaqueous phase liquid contaminants and has not been extended to chemicals other than hydrocarbons. In contrast to

INTRODUCTION There is often an oxic/anoxic interface between the vadose zone and groundwater polluted with organic contaminants. The aerobic biodegradation of petroleum hydrocarbons in the vadose zone has been well established.1,2 Diffusion of oxygen from the surface and volatilization of petroleum hydrocarbons that can serve as electron donors support the microbes that create a redox gradient in the vadose zone. The metabolic rates and biomass in the vadose zone increase in proportion to the concentration of the limiting nutrient.1,3 Key parameters including moisture,4 temperature,5 pH,5 substrate, nutrient, and electron acceptor availability6 also affect microbial growth. Moisture plays a particularly important role for microbes in the vadose zone: microbial growth and activity rapidly decrease with depth and increase again in the capillary fringe.3,7 The capillary fringe is located in the unsaturated layer of subsurface solids above the water table where air and water intersect. The oxic/anoxic interface in the vadose zone of many contaminated sites often occurs close to the capillary fringe.8 It is in principle similar to contaminated plume fringes9−11 in that it is an interface with steep redox gradients where electron donors or acceptors are available and high microbial activity is present.12,13 Biodegradation within contaminant plumes has been studied extensively. Stable carbon isotope fractionation and/or tracking of functional genes involved in the biodegradation pathways have been used to determine hydrocarbon, methane, and toluene biodegradation at the © 2013 American Chemical Society

Special Issue: Rene Schwarzenbach Tribute Received: Revised: Accepted: Published: 6846

December 3, 2012 February 21, 2013 March 8, 2013 March 8, 2013 dx.doi.org/10.1021/es3049465 | Environ. Sci. Technol. 2013, 47, 6846−6854

Environmental Science & Technology

Article

min−1 for 5 min. Chlorobenzenes were monitored at 260 nm with a detection limit of 0.01 mg • min−1. Soil samples were extracted twice with acetonitrile/water (1:1) prior to analysis. All the samples were centrifuged for 1 min prior to HPLC analysis. Samples Used for the Study. Preliminary experiments were conducted using sand as the column packing material. Subsequent studies were done with subsurface solids from a contaminated field site. Samples were collected The DuPont Chambers Works in New Jersey in June, 2011. Sediment cores (5 cm diameter) were collected using a drill rig above a groundwater plume contaminated with CB, 12DCB, and 14DCB (10 mg • kg−1, 5 mg • kg−1, and 5 mg • kg−1). The water table was 76 cm below ground surface when the samples were collected. Physical and chemical parameters of the vadose zone were obtained by analyzing composite samples every 7.5 cm over the length of the cores. Microcosms. Microcosms were prepared with samples from the cores to evaluate potential biodegradation activity in the vadose zone. Solids (10 g) were mixed with 50 mL of 1/4strength Stanier’s mineral salts medium (MSB)34 containing CB (10 mg • L−1), 12DCB (5 mg • L−1), and 14DCB (5 mg • L−1) in 160-mL serum bottles sealed with Teflon lined stoppers. Additional chlorobenzenes were added with a syringe when necessary. At appropriate intervals, headspace samples were analyzed by GC to determine the concentrations of CB, 12DCB, and 14DCB. Degradation rates were estimated after the lag period. Column Design. A 30 × 2.2 cm (I.D.) multiport Chromaflex glass column (Figure 1) was packed with sand or

petroleum hydrocarbons, chlorinated aliphatic solvents, including perchloroethylene and trichloroethylene, typically are not biodegraded under aerobic conditions in the vadose zone. They can migrate substantial distances and cause vapor intrusion in buildings.22−24 Chlorinated aromatic compounds contribute to groundwater contamination through their use as synthetic intermediates and as solvents. Chlorobenzene (CB) and 1,2-dichlorobenzene (12DCB) are industrial solvents, and 1,4-dichlorobenzene (14DCB), a possibile carcinogen, is a commonly used volatile disinfectant and deodorizer.25 It is well established that CB, 12DCB, and 14DCB undergo aerobic biodegradation,26−29 but they are more resistant to biodegradation under anaerobic conditions.27,30,31 Chlorobenzenes are volatile; therefore, they have the potential to enter buildings as vapors from contaminant plumes in groundwater. Aerobic biodegradation in the vadose zone, however, could prevent transport in the vadose zone if the chlorobenzenes behave like hydrocarbons rather than like chlorinated aliphatic solvents. Bacterial biomass can be highly concentrated in a thin layer where gradients of electron donors and acceptors intersect. For example, in microbial mats32 or in oxygen minimum zones,33 the active microbes increase in number until the electron donor or acceptor becomes limiting. We recently found high rates of chlorobenzene biodegradation in a 2 mm thick sediment/water interface.11 The same principle seems likely to apply at the interface in the vadose zone where contaminants act as electron donors and oxygen is the electron acceptor for aerobic biodegradation. Therefore we hypothesized that aerobic degradation near the capillary fringe in the vadose zone could result in an active bacterial population at the oxic/anoxic interface with a sufficient capacity to protect the overlying air.



MATERIALS AND METHODS Chemicals. Chlorobenzene (CB) (99.5%), 1,2-dichlorobenzene (99%), and 1,4-dichlorobenzene (≥99%) were from Sigma-Aldrich. Resazurin sodium salt was from Acros Organics, N.V. Analytical Methods. Gas samples (100 μL) were analyzed with an Agilent 6890N Gas Chromatograph (GC) equipped with a flame ionization detector (FID) and a Supelco 10% 80/ 100 Chromosorb W HP column to measure the concentrations of CB, 12DCB, and 14DCB with minimum detection limit of 0.05 mg of contaminant • mL−1 of air. The gas calibrations were performed by injection of headspace samples from 160 mL glass serum vials sealed with Teflon lined stoppers containing 50 mL of sterile deionized water and measured amounts of chlorinated benzenes. Calculations were based on the Henry’s law constants of the chlorinated benzenes at 23 °C. The oven temperature was 120 °C, and the gas flow rates were 45 mL of air • min−1, 40 mL of hydrogen • min−1, and 25 mL of helium • min−1. Oxygen concentrations were measured by using an Agilent 6850 GC equipped with a thermal conductivity detector (TCD) and an HP-PLOT MoleSieve column (30 m × 0.53 mm × 25 μm). The detector temperature was 250 °C, and the oven temperature was 50 °C with helium as the carrier gas at a flow rate of 3.3 mL • min−1. Duplicate samples were analyzed for each measurement. CB, 12DCB, and 14DCB concentrations in liquid samples were analyzed by high-performance liquid chromatography (HPLC) using a Merck Chromolith RP18e column (4.6 × 100 mm) with a mobile phase of 40:60 water/acetonitrile (with acetonitrile containing 0.05% trifluoroacetic acid) at a flow rate of 1 mL •

Figure 1. Schematic diagram of the vadose zone column and continuous flow system. U-tubes on the outlets prevented oxygen diffusion into the column from the effluent ports and glass beads occupied 2 cm of column height.

subsurface solids and kept at room temperature (22−24 °C). An aerobic zone was established/maintained by passing humidified and filter sterilized (0.2 μm filter) air across the top of the column at a rate of 1 mL • h−1 with a peristaltic pump. Chlorobenzenes entered the column via volatilization from the filter sterilized (0.2 μm Teflon filter) anoxic water pumped through the bottom of the column with a syringe pump at a flow rate of 1 mL • h−1. Effluent water that was trapped in U tubes prevented diffusion of atmospheric oxygen through the effluent port back into the groundwater in the 6847

dx.doi.org/10.1021/es3049465 | Environ. Sci. Technol. 2013, 47, 6846−6854

Environmental Science & Technology

Article

Table 1. Bacteria Used To Inoculate Columns

a

isolate

closest relative

identitya

accession no.

JS691 JS692 JS694 JS696 JS697 JS698 JS699 JS782 JS783 JS784

Stenotrophomonas sp. B3a Stenotrophomonas maltophilia strain YSP48 Pseudomonas pseudoalcaligenes strain 2-3 Pseudomonas stutzeri OC-10 Pseudomonas sp. 1410 Pseudomonas stutzeri strain A5 Pseudomonas stutzeri strain K-2-7 Pseudomonas sp. MM15 Pseudomonas stutzeri strain NCG1 Pseudomonas sp. QHL22

99% 99% 99% 98% 99% 99% 100% 99% 99% 100%

GU394955 JF894170 HQ285878 AY669170.1 JN645996.1 JN613328.1 JQ963329.1 JF419326.1 JN712254.1 JQ860237.1

growth substrates CB CB CB 12DCB, 12DCB 12DCB, 14DCB, 14DCB, CB CB

ref 11 11 11

CB CB CB CB

this this this this this this this

study study study study study study study

Based on 16S rRNA gene sequences (800 bp).

mm glass beads. The lower 2 cm of the column was wrapped with stainless steel screen (0.178 mm mesh) to prevent flow of the solids into the outlet tubes. A control experiment under anoxic conditions was performed by pumping nitrogen instead of air across the top of the column at a rate of 5 mL • h−1. Most-Probable-Number Analysis. Samples were homogenized by brief (3 times of 5 s) treatment in a bead beater (BioSpec Mini-BeadBeater) at maximum speed using 2.5-mm zirconia/silica beads prior to enumeration of bacteria by mostprobable-number (MPN) estimation. For the sand column MPN determinations, serial dilutions were prepared in 1/4strength MSB containing bromothymol blue (20 mg/L) in 96well microplates. The plates were incubated in desiccators with CB provided in the headspace. Cultures positive for CB degradation were indicated by a color change due to CB degradation and release of HCl.28 The microbial count was calculated from an 8-tube MPN table with 95% confidence limits.38 Bacteria in the samples obtained from the column filled with site material were quantified by spreading 100 μL of appropriately diluted homogenized samples on 1/4-strength MSB agar plates (in duplicate) incubated in desiccators with CB, 12DCB, or 14DCB in the headspace. Quantitative Polymerase Chain Reaction (qPCR). DNA was extracted with a PowerSoil DNA Isolation Kit (MO BIO Laboratories, Carlsbad, CA). qPCR was done with an established protocol39 using 2X Power SYBR Green PCR Master Mix (Applied Biosystems, Carlsbad, CA) and an ABI 7500 Fast Real-Time PCR System equipped with SDS v. 2.0.3 software using the default SYBR Green cycling parameters. Three sets of primers were used to determine total bacteria, isolated degraders, and chlorobenzene dioxygenase for triplicate samples. BacF/BacR primers were used for total bacterial 16S rRNA genes,40 and TOD-F/TOD-R primers were used for chlorobenzene dioxygenase39 gene quantification. The 5′GAGGTAAAGGCTCACCAAGG-3′ forward and 5′TCTGGACCGTGTCTCAGTTC-3′ reverse primers were designed to estimate the total biomass of the inoculum based on a region of the 16S rRNA gene sequence common to all the isolates capable of degrading CB, 12DCB, or 14DCB (Table 1). Calibrations41 were performed in triplicate with serial 10-fold dilutions of the 16S rRNA gene of Pseudomonas sp. strain JS15026 for the total and isolated bacteria. The chlorobenzene dioxygenase gene of Ralstonia sp. strain JS70542 was cloned using a pGEM-T Easy Vector System and used as standard for chlorobenzene dioxygenase. Standard curves gave slopes of 3.2−3.4 and R2 values higher than 0.90. It was assumed that there was one chlorobenzene dioxygenase copy per cell, 4.1

column. The feed was prepared under nitrogen, and the reduction potential was monitored by the inclusion of resazurin (2 mg • L−1) and following blue color. Stainless steel tubing was used for all connections. The oxygen and contaminant concentrations were monitored by sampling at 6 ports spaced at the indicated intervals over the length of the column. The sand column was packed with 196 g of sand above 8.3 g of 3mm glass beads using the procedure described by Oliveira et al.35 The sand used for the column was sieved (0.6 to 0.42 mm sieve size) and baked at 550 °C overnight. A stainless steelmesh (0.178 mm mesh) screen was wrapped inside the column to prevent sand particles from blocking the sampling ports. Using the equations of Atherton et al. the capillary fringe thickness in the column was estimated to be between 12 and 17 cm above the saturated zone 36 which was consistent with visual observations. The sand column was inoculated with 20 mL of a culture suspension containing equal volumes of the CBdegrading bacteria JS691, JS692, and JS694 (Table 1). The strains were previously isolated from the sediment/water interface of a canal adjacent to the manufacturing site where the core samples were collected. The cultures were grown with CB as the carbon source,28 washed twice by centrifugation with 1/4-strength MSB, and suspended to an OD600 of 0.6 in 1/4strength MSB. The total porosity in the column was measured as 50%; therefore, the seepage velocity in the column was calculated to be 46 m • h−1. As a control for adsorptive or diffusive losses the column was operated abiotically until CB and oxygen concentrations were stable (no change in concentration gradients) for 5 days prior to inoculation. When field samples were used to fill the column, 133 g of vadose zone solids was added to the column on top of 8 g of 3mm glass beads. Because the cores consisted of heterogeneous sandy clay, the column fill was prepared by combining vadose zone core samples from 12 cm above to 12 cm below the capillary fringe and sieving the material through a size 0.853 mm sieve.37 The porosity of the material was measured as 35%, and the seepage velocity in the column was calculated to be 51 m • h−1. The moisture content of the mixture was adjusted to 15% by weight prior to sieving by inoculating it with a suspension of equal volumes of CB-, 12DCB-, and 14DCBdegrading bacteria isolated from the microcosms (Table 1, Figure 5) and grown in 1/4-strength MSB supplemented with the appropriate chlorobenzene isomer. The bacterial culture was washed once by centrifugation with 1/4-strength MSB and suspended to an OD600 of 0.02 in 1/4-strength MSB. The inoculum was lower than in the sand column to be more consistent with field conditions. The column was packed by filling and tamping the prepared site material above 8.3 g of 36848

dx.doi.org/10.1021/es3049465 | Environ. Sci. Technol. 2013, 47, 6846−6854

Environmental Science & Technology

Article

Figure 2. Influence of airflow and CB influent concentrations in sterile sand column. Nominal CB feed concentrations were 10 mg • L−1 (A and C) or 30 mg • L−1 (B and D). Air (1 mL • h−1) (A and C) or no air (B and D) was pumped across the surface of the column. Columns were allowed to equilibrate for 7 days after each change in conditions. Here and throughout the data represent means of duplicate analyses.

Figure 3. Biodegradation in inoculated sand columns. CB and oxygen profiles 120 h after inoculation of the column operated with 10.5 mg of CB • L−1 (A) and 96 h after switching the feed concentration of CB to 31.1 mg of CB • L−1 (B).

resulted in effluent concentrations of 9.6 ± 0.6 and 30.1 ± 1.1 mg of CB • L−1 in the liquid when air was pumped over the column at a rate of 1 mL • h−1. Similarly, influent concentrations of 10.2 ± 0.7 and 30.5 ± 1.5 mg of CB • L−1 resulted in effluent concentrations of 9.7 ± 0.3 and 30.2 ± 2.1 mg of CB • L−1 when no air was pumped across the column. Pumping air across the top of the column at 1 mL • h−1 did not have a significant effect on oxygen or CB profiles (Figure 2). Biodegradation was evaluated by monitoring the CB and oxygen profiles after the concentrations became stable 120 h after inoculation (Figure 3). CB was removed to below the detection limits in the capillary fringe less than 5 cm above the saturated zone when the anoxic feed was 10 mg of CB • L−1. The column was operated for 25 days with no change in behavior (Figure S2), which indicated that the system was robust. The differences in transport of oxygen and contaminants in the capillary fringe and in the overlying vadose zone

copies of the 16S rRNA gene per cell,43 and 6 copies of the 16S rRNA gene per Pseudomonas cell.44



RESULTS AND DISCUSSION Biodegradation in Sand Columns. Biodegradation of volatile contaminants from air by microbes has been widely studied in gas phase reactors where microbial cultures are kept in liquid and contaminants are provided as a vapor.45−48 CB biodegradation in the vapor phase has also been established using biofilters and biotrickling filters49−51 to treat contaminated gases. In this study preliminary experiments were performed using sand columns to determine whether bacteria capable of degrading chlorinated compounds could survive and biodegrade the contaminants in the vadose zone. Sterile columns were operated until the oxygen and CB profiles were stable for at least 5 days (Figure S1). Influent concentrations of 10.3 ± 0.5 and 30.7 ± 1.8 mg of CB • L−1 6849

dx.doi.org/10.1021/es3049465 | Environ. Sci. Technol. 2013, 47, 6846−6854

Environmental Science & Technology

Article

gave rise to profiles in which the slopes were different above and below the top of the capillary fringe. The column was washed with 20 mL of 1/100-strength air saturated MSB pumped onto the top of the column at a flow rate of 1 mL • h−1 to evaluate the potential effects of rainfall (Figure S3). Within 140 h after washing, the oxygen and CB profiles were indistinguishable from the prewash profiles, which indicated that the bacteria adhered well to the sand in the column. The oxygen limitation was evaluated by increasing the CB concentration in the feed to 30 mg of CB • L−1. At the higher concentration CB migrated further into the vadose zone before sufficient oxygen was available to support biodegradation (Figure 3). The system was stable during operation for 10 days. In all of the biologically active conditions the CB concentrations immediately above the saturated zone were substantially lower (Figure 3) than in the controls (Figure 2). The effect would steepen the diffusion gradient and thus increase the migration of contaminants from the water.52 The difference in flux between active (69 mg of CB • m−2 • d−1) and control (44 mg of CB • m−2 • d−1) conditions provides an estimate of the effect. The biodegradation rate of 25 ± 3 mg of CB • m−2 • d−1 in the sand columns was calculated from the concentration difference between influent and effluent in the liquid when the anoxic feed contained 10 mg of CB • L−1. The rates were consistent with estimates obtained by comparing the gas phase concentrations in sterile and active systems integrated by column height (data not shown). CB contamination in the groundwater at the field site varies from 1.6−15 mg CB • L−1;53 therefore, the biodegradation in the unsaturated sand column suggested that there is a potential for the microbes in the field to prevent vadose zone contamination via natural attenuation over a range of CB plume concentrations. Total microbial biomass in the vadose zone increases with increasing biodegradation.54 To determine the distribution of biomass in the vadose zone column, the active microbial biomass was measured along the length of the column by MPN and qPCR assays at the end of the experiment. The 16S rRNA gene sequences of total bacteria (BacF/BacR primers) measured the total microbial count, while amplification with TodF/TodR primers for chlorobenzene dioxygenase gene sequences indicated the number of CB degrading bacteria. The results indicated that the majority of the active microbes in the unsaturated sand were found in the first 10 cm above the water table in the upper part of the capillary fringe (Figure 4) which coincides with the area where CB disappeared. Biodegradation in the Column Filled with Subsurface Solids. The subsurface solids from the core samples were highly heterogeneous with alternating layers of clay, sandy clay, and gravel. Chlorobenzenes were detected in the cores, but the heterogeneity of the cores and volatility of the contaminants precluded accurate profiling of the chlorobenzenes. The presence of chlorobenzene degrading bacteria and their potential activities in the vadose zone were evaluated by constructing microcosms with samples of the solids and calculating biodegradation rates after the lag periods (Figure 5-A). Similar results were obtained with samples from a second core from the same area (data not shown). Estimates of microbial activity were consistent with initial bacterial counts and moisture content (Figure 5). The preliminary observations provided evidence of active chlorobenzene degrading bacteria

Figure 4. Distribution of bacteria in the sand column after 35 days of operation. Estimates of bacteria based on MPN with CB ● (light blue), 16S rRNA genes ● (light green), toluene/CB dioxygenase genes ● (red). Dashed line represents the average initial bacterial inoculum estimated by quantifying 16S rRNA genes.

Figure 5. Analysis of vadose zone samples. (A) Biodegradation of CB ● (red), 12DCB ● (purple), and 14DCB ● (light green) in microcosms prepared with vadose zone solids. (B) pH △, moisture content ▲, and MPN of CB ● (red), 12DCB ● (gray), and 14DCB ● (light green) degrading bacteria were determined in samples of the vadose zone core.

and ongoing natural attenuation of chlorobenzenes in the vadose zone but provided no indication of rates or fluxes. The multiport column was packed with subsurface solids from the cores to evaluate the potential for biodegradation in the vadose zone. The goal was to determine whether biodegradation could prevent migration of chlorobenzenes to the vadose zone rather than to directly predict biodegradation rates at the field site. Therefore, the moisture content, particle size, and pH of the solids were adjusted, and chlorobenzene degrading bacteria previously isolated from the core samples were added prior to constructing the columns. The protocol provided sufficient bacteria, void volume for substrate transport, and buffering capacity for microbial growth.55,56 CB, 12DCB, and 14DCB were biodegraded in the column within 144 h when the anoxic feed contained CB (10 mg • 6850

dx.doi.org/10.1021/es3049465 | Environ. Sci. Technol. 2013, 47, 6846−6854

Environmental Science & Technology

Article

L−1), 12DCB (5 mg • L−1), and 14DCB (5 mg • L−1) (Figure S4). Concentration profiles were stable for 168 h (Figure 6-

with visual observations in the column. Because of the way the solids were prepared the pore spaces were larger than would be expected in the field. Therefore, bulk diffusion rates in the column were likely to be higher, and the height of the capillary fringe was likely to be lower than under field conditions. Bacteria associated with the solids in the column were determined by quantifying the MPNs, 16S rRNA genes, and toluene/CB dioxygenase sequences at the end of the 28-day experiment. As in the sand column, the highest estimates of culturable CB degrading bacteria and 16S rRNA genes of total bacteria occurred at the top of the capillary fringe (Figure 7).

Figure 6. Transport and biodegradation of chlorobenzenes in the column filled with subsurface solids from the field site. Column operated with (A) air after 312 h, (B) nitrogen instead of air for 86 h, and (C) the same column 72 h after reintroduction of air.

Figure 7. Enumeration of bacteria in the column filled with subsurface solids after 28 days of operation. (A) Moisture content ▲ and MPNs of CB ● (red), 12DCB ● (purple), and 14DCB ● (light green) degraders. (B) qPCR estimates of final numbers of CB degraders ● (red), inoculated degraders ● (green), and total bacteria ● (blue). Dashed lines represent the average initial count of CB degraders (red), inoculated degraders (green), and total bacteria (blue) in the column fill material.

A).The biodegradation in the column stopped within 48 h after flow was switched to nitrogen, and the contaminant concentrations remained stable for 38 h (Figure 6-B). The biodegradation in the column resumed within 72 h after air was introduced (Figure 6-C) and continued with no change in oxygen or contaminant profiles for 210 h. The results indicate that the biodegradation was robust, oxygen dependent, and concentrated within the capillary fringe. Based on the difference between influent and effluent liquid concentrations in the column the biodegradation rates (fluxes) were 21 ± 1 mg of CB • m−2 • d−1, 3.7 ± 0.5 mg of 12DCB • m−2 • d−1, and 7.4 ± 0.7 mg of 14DCB • m−2 • d−1. The rates were similar to those in the sand column indicating that the natural field sediments could support biodegradation of chlorobenzenes. The differences between the influent and effluent concentrations of the contaminants in the anoxic feed were 0.4, 0.1, and 0.3 mg • L−1 for CB, 12DCB, and 14DCB, during the control phase, and 1, 0.3, and 0.8 mg • L−1 for CB, 12DCB, and 14DCB, during the active phase. The results are consistent with the idea, supported by Fick’s first law, that biodegradation in the vadose zone can increase the flux of volatile biodegradable contaminants and thereby decrease the contaminant concentrations in anoxic plumes.57 The capillary fringe in the column prepared with site material was predicted to extend 8 cm above the saturated zone based on the particle diameter of 0.841 mm36 which was consistent

The MPN and total 16S rRNA gene counts were lower than in the sand column because the sand column received a larger inoculum and was operated for a longer time. In addition, the bacteria may have been more difficult to dislodge from the authentic aquifer materials. The 14DCB degraders and some 12DCB degraders used to inoculate the column were capable of degrading CB, whereas the remaining isolates were restricted to single substrates (Table 1). The estimates based on 16S rRNA genes were very similar to the MPN counts and indicated the location of maximum bacterial biomass in the column (Figure 7-B). The CB dioxygenase gene encodes the enzymes that initiate, CB, 12DCB, and 14DCB biodegradation.58,59 The CB dioxygenase gene counts were consistent with the MPNs and enumeration based on 16S rRNA genes of the inoculated bacteria (Figure 7-B). The primers for CB dioxygenase worked well with the isolates used as the inoculum, but their effectiveness with other indigenous bacteria able to degrade chlorobenzenes is unknown. Therefore, the numbers in Figure 7 are conservative estimates of the guilds involved in the degradation processes. The higher numbers of total bacteria are consistent with the presence of other chlorobenzene degraders as well as other 6851

dx.doi.org/10.1021/es3049465 | Environ. Sci. Technol. 2013, 47, 6846−6854

Environmental Science & Technology

Article

transport of contaminant and electron acceptor. The biodegradation there is so fast as to be best described by the instantaneous model discussed most recently by Davis et al.8 We applied the instantaneous model8,9 using the diffusion coefficients given in the Ettinger and Johnson model63 and oxygen stoichiometry (1.04 mg of oxygen/mg of CB, 1.06 mg of oxygen/mg of 12DCB or 14DCB) estimated by assuming that 50% of the carbon was converted to biomass (Figure S5). The fit was much better and consistent with the idea that the biodegradation is controlled only by the flux of oxygen and contaminants. The simple model used by Davis was designed to describe transport and biodegradation in the vadose zone above the capillary fringe and could perhaps be modified to describe biodegradation within the capillary fringe. Although the current studies clearly establish the potential for biodegradation in the vadose zone, it is difficult to describe the complicated factors controlling the kinetics in the capillary fringe. In addition, field validation must be conducted as has been done for petroleum hydrocarbons.8 Finally, evaluation and prediction of biodegradation in the field will require application of a model that includes oxygen, substrate- and organism-dependent biodegradation kinetics along with the type and physical parameters of the vadose zone solids with particular emphasis on the capillary fringe.21,37,55,62

members of the indigenous microbial community not directly involved in biodegradation of chlorobenzenes. Taken together the microbial enumeration results revealed that the majority of the active microbes in the vadose zone, as in other oxic/anoxic interfaces,60,61 are concentrated in relatively narrow zones characterized by high flux and steep gradients of electron donors and acceptors. The biomass in the narrow active zone seems to increase in proportion to the flux rather than the concentrations of CB and oxygen. The modest number of sampling ports led to limited resolution of the oxygen and chlorobenzene gradients in the columns. The contaminant concentration profiles in the columns were measured every 5 cm, whereas the biomass concentration was measured every 2 cm. As a consequence, the highest biomass concentrations were found slightly above the levels where contaminant concentrations became undetectable. The location of highest microbial activity would be dependent on moisture content and availability of contaminants and oxygen. Conversely, the area of highest activity would create the steep gradients of contaminants and oxygen concentrations so that the areas of highest flux would correspond to areas of lowest concentrations. Although the above limitations make it difficult to measure the relationships among contaminants, oxygen and biomass at fine scales, the results clearly establish that chlorobenzenes volatilizing from contaminated water are readily biodegraded in the lower part of vadose zone and that the degradation was limited only by transport of contaminants and oxygen. The rate and extent of natural attenuation of petroleum hydrocarbons in the vadose zone is controlled by oxygen availability and moisture content because hydrocarbon degrading bacteria are ubiquitous. Natural attenuation of other contaminants may be more complex. Dissolved volatile chlorinated compounds that are anaerobically recalcitrant will evaporate from subsurface plumes and can cause vapor intrusion into buildings. PCE and TCE are resistant to aerobic biodegradation and so are transported through the vadose zone with little attenuation.22−24 The results of this study revealed remarkable biodegradation capacities for chlorinated aromatic compounds in a very narrow section of the capillary fringe of the columns containing solids and bacteria from the vadose zone. Thus at least under laboratory conditions the chlorobenzenes behave more like petroleum hydrocarbons than like PCE and TCE. It seems likely that the findings will be applicable to a variety of volatile synthetic contaminants that are resistant to anaerobic biodegradation but degradable under aerobic conditions. The key requirements would be the presence of appropriate bacteria along with sufficient moisture and oxygen. The oxygen demand under field conditions would also include natural organic matter which was minimized for the purposes of the experiments described here. Our results illustrate a key problem with the use of first order biodegradation rates in modeling contaminant transport. BioVapor software22 was used to model the measured concentration by varying the first order biodegradation constants. The constants obtained (106 h−1 for CB, 35 h−1 for 12DCB and 65 h−1 for 14DCB) were far outside the range obtained in the literature or the constants obtained from microcosms prepared with site material (0.021 h−1 for CB, 0.003 h−1 for 12DCB, and 0.009 h−1 for 14DCB), and the results would be misleading if extrapolated to the entire vadose zone. Here and likely in the field the biodegradation takes place in a very narrow zone and appears to be limited only by the



ASSOCIATED CONTENT

S Supporting Information *

Figures S1−S5. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Phone: (404) 894-0628. Fax: (404) 894-8266. E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS Funding was provided by DuPont Corporate Remediation Group. We thank the sampling team from URS for collecting the samples. The authors thank Shirley Nishino and E. Erin Mack for reviewing the manuscript and help throughout the study.



REFERENCES

(1) Holden, P. A.; Fierer, N. Microbial processes in the vadose zone. Vadose Zone J. 2005, 4 (1), 1−21. (2) Fisher, J. M.; Baker, R. J.; Matthew, A. L.; Baehr, A. L. Determination of vapor phase diffusion coefficients for unsaturated zone sediments at a gasoline spill site in Galloway Township, New Jersey. Water- Resour. Invest. Rep. (U. S. Geol. Surv.) 1993, 94, 35−41. (3) Konopka, A.; Turco, R. Biodegradation of organic compounds in vadose zone and aquifer sediments. Appl. Environ. Microbiol. 1991, 57 (8), 2260−2268. (4) Wang, B.; Zhao, Y.-s.; Qu, Z.-h.; Zheng, W.; Zhu, W.; Long, B.-s.; Jiao, L.-n.; Xu, C. Impact of depth and moisture to diesel degradation in sand layer of vadose zone. Huanjing Kexue 2011, 32 (2), 530−535. (5) Antizar-Ladislao, B. Bioremediation: working with bacteria. Elements 2010, 6 (6), 389−394. (6) Molins, S.; Mayer, K. U.; Amos, R. T.; Bekins, B. A. Vadose zone attenuation of organic compounds at a crude oil spill site - Interactions between biogeochemical reactions and multicomponent gas transport. J. Contam. Hydrol. 2010, 112 (1−4), 15−29.

6852

dx.doi.org/10.1021/es3049465 | Environ. Sci. Technol. 2013, 47, 6846−6854

Environmental Science & Technology

Article

(7) Taylor, J. P.; Wilson, B.; Mills, M. S.; Burns, R. G. Comparison of microbial numbers and enzymatic activities in surface soils and subsoils using various techniques. Soil Biol. Biochem. 2002, 34 (3), 387−401. (8) Davis, G. B.; Patterson, B. M.; Trefry, M. G. Evidence for instantaneous oxygen-limited biodegradation of petroleum hydrocarbon vapors in the subsurface. Ground Water Monit. Rem. 2009, 29 (1), 126−137. (9) Chu, M.; Kitanidis, P. K.; McCarty, P. L. Modeling microbial reactions at the plume fringe subject to transverse mixing in porous media: When can the rates of microbial reaction be assumed to be instantaneous? Water Resour. Res. 2005, 41, (6). (10) Vieth, A.; Kastner, M.; Schirmer, M.; Weiss, H.; Godeke, S.; Meckenstock, R. U.; Richnow, H. H. Monitoring in situ biodegradation of benzene and toluene by stable carbon isotope fractionation. Environ. Toxicol. Chem. 2005, 24 (1), 51−60. (11) Kurt, Z.; Shin, K.; Spain, J. Biodegradation of chlorobenzene and nitrobenzene at interfaces between sediment and water. Environ. Sci. Technol. 2012, 46, 11829−11835. (12) Winderl, C.; Anneser, B.; Griebler, C.; Meckenstock, R. U.; Lueders, T. Depth-resolved quantification of anaerobic toluene degraders and aquifer microbial community patterns in distinct redox zones of a tar oil contaminant plume. Appl. Environ. Microbiol. 2008, 74 (3), 792−801. (13) Bauer, R. D.; Maloszewski, P.; Zhang, Y.; Meckenstock, R. U.; Griebler, C. Mixing-controlled biodegradation in a toluene plume results from two-dimensional laboratory experiments. J. Contam. Hydrol. 2008, 96 (1−4), 150−168. (14) Alfreider, A.; Vogt, C. Bacterial diversity and aerobic biodegradation potential in a BTEX-contaminated aquifer. Water, Air, Soil Pollut. 2007, 183 (1−4), 415−426. (15) Amos, R. T.; Bekins, B. A.; Delin, G. N.; Cozzarelli, I. M.; Blowes, D. W.; Kirshtein, J. D. Methane oxidation in a crude oil contaminated aquifer: Delineation of aerobic reactions at the plume fringes. J. Contam. Hydrol. 2011, 125 (1−4), 13−25. (16) Roggemans, S.; Bruce, C. L.; Johnson, P. C., Vadose zone natural attenuation of hydrocarbon vapors: an empirical assessment of soil gas vertical profile data. In American Petroleum Institute; API Technical Bulletin: Washington, DC, 2002; Vol. 15. (17) Franzmann, P. D.; Zappia, L. R.; Power, T. R.; Davis, G. B.; Patterson, B. M. Microbial mineralisation of benzene and characterisation of microbial biomass in soil above hydrocarbon-contaminated groundwater. FEMS Microbiol. Ecol. 1999, 30 (1), 67−76. (18) Abreu, L. D. V.; Johnson, P. C. Effect of vapor source - building separation and building construction on soil vapor intrusion as studied with a three-dimensional numerical model. Environ. Sci. Technol. 2005, 39 (12), 4550−4561. (19) Abreu, L. D. V.; Johnson, P. C. Simulating the effect of aerobic biodegradation on soil vapor intrusion into buildings: Influence of degradation rate, source concentration, and depth. Environ. Sci. Technol. 2006, 40 (7), 2304−2315. (20) Devaull, G. E. Indoor vapor intrusion with oxygen-limited biodegradation for a subsurface gasoline source. Environ. Sci. Technol. 2007, 41 (9), 3241−3248. (21) Ma, J.; Rixey, W. G.; Devaull, G. E.; Stafford, B. P.; Alvarez, P. J. J. Methane bioattenuation and implications for explosion risk reduction along the groundwater to soil surface pathway above a plume of dissolved ethanol. Environ. Sci. Technol. 2012, 46 (11), 6013−6019. (22) Forand, S. P.; Lewis-Michl, E. L.; Gomez, M. I. Adverse birth outcomes and maternal exposure to trichloroethylene and tetrachloroethylene through soil vapor intrusion in New York state. Environ. Health Perspect. 2012, 120 (4), 616−621. (23) McDonald, G. J.; Wertz, W. E. PCE, TCE, and TCA vapors in subslab soil gas and indoor air: A case study in upstate New York. Ground Water Monit. Rem. 2007, 27 (4), 86−92. (24) Office of Underground Storage Tanks Petroleum hydrocarbons and chlorinated hydrocarbons differ in their potential for vapor intrusion; U.S. Environmental Protection Agency: WA, 2012; pp 1−13.

(25) Djohan, D.; Yu, J.; Connell, D.; Christensen, E. Health risk assessment of chlorobenzenes in the air of residential houses using probabilistic techniques. J. Toxicol. Environ. Health 2007, 70 (19), 1594−1603. (26) Haigler, B. E.; Nishino, S. F.; Spain, J. C. Degradation of 1,2 dichlorobenzene by a Pseudomonas sp. Appl. Environ. Microbiol. 1988, 54 (2), 294−301. (27) Spain, J. C.; Nishino, S. F. Degradation of 1,4-dichlorobenzene by a Pseudomonas sp. Appl. Environ. Microbiol. 1987, 53 (5), 1010− 1019. (28) Nishino, S. F.; Spain, J. C.; Belcher, L. A.; Litchfield, C. D. Chlorobenzene degradation by bacteria isolated from contaminated groundwater. Appl. Environ. Microbiol. 1992, 58 (5), 1719−1726. (29) Nishino, S. F.; Spain, J. C.; Pettigrew, C. A. Biodegradation of chlorobenzene by indigenous bacteria. Environ. Toxicol. Chem. 1994, 13 (6), 871−877. (30) Fung, J. M.; Weisenstein, B. P.; Mack, E. E.; Vidumsky, J. E.; Ei, T. A.; Zinder, S. H. Reductive dehalogenation of dichlorobenzenes and monochlorobenzene to benzene in microcosms. Environ. Sci. Technol. 2009, 43 (7), 2302−2307. (31) Dilmeghani, M.; Zahir, K. O. Kinetics and mechanism of chlorobenzene degradation in aqueous samples using advanced oxidation processes. J. Environ. Qual. 2001, 30 (6), 2062−2070. (32) Krueger, M.; Blumenberg, M.; Kasten, S.; Wieland, A.; Kaenel, L.; Klock, J.-H.; Michaelis, W.; Seifert, R. A novel, multi-layered methanotrophic microbial mat system growing on the sediment of the Black Sea. Environ. Microbiol. 2008, 10 (8), 1934−1947. (33) Stewart, F. J.; Ulloa, O.; DeLong, E. F. Microbial metatranscriptomics in a permanent marine oxygen minimum zone. Environ. Microbiol. 2012, 14 (1), 23−40. (34) Behrman, E. J.; Stanier, R. Y. The bacterial oxidation of nicotinic acid. J. Biol. Chem. 1957, 228, 923−945. (35) Oliveira, I. B.; Demond, A. H.; Salehzadeh, A. Packing of sands for the production of homogeneous porous media. Soil. Sci. Soc. Am. J. 1996, 60 (1), 49−53. (36) Atherton, R. J.; Baird, A. J.; Wiggs, G. F. S. Inter-tidal dynamics of surface moisture content on a meso-tidal beach. J. Coast. Res. 2001, 17 (2), 482−489. (37) Kristensen, A. H.; Hosoi, C.; Henriksen, K.; Loll, P.; Moldrup, P. Vadose zone biodegradation of benzene vapors in repacked and undisturbed soil cores. Vadose Zone J. 2012, 11, (1). (38) Garthright, W. E.; Blodgett, R. J. FDA’s preferred MPN methods for standard, large or unusual tests, with a spreadsheet. Food Microbiol. 2003, 20 (4), 439−445. (39) Baldwin, B. R.; Nakatsu, C. H.; Nies, L. Detection and enumeration of aromatic oxygenase genes by multiplex and real-time PCR. Appl. Environ. Microbiol. 2003, 69 (6), 3350−3358. (40) Chung, J.; Krajmalnik-Brown, R.; Rittmann, B. E. Bioreduction of trichloroethene using a hydrogen-based membrane biofilm reactor. Environ. Sci. Technol. 2008, 42 (2), 477−483. (41) Whelan, J. A.; Russell, N. B.; Whelan, M. A. A method for the absolute quantification of cDNA using real-time PCR. J. Immunol. Methods 2003, 278 (1−2), 261−269. (42) van der Meer, J. R.; Werlen, C.; Nishino, S. F.; Spain, J. C. Evolution of a pathway for chlorobenzene metabolism leads to natural attenuation in contaminated groundwater. Appl. Environ. Microbiol. 1998, 64 (11), 4185−4193. (43) Lee, Z. M.-P.; Bussema, C., III; Schmidt, T. M. rrnDB: documenting the number of rRNA and tRNA genes in bacteria and archaea. Nucleic Acids Res. 2009, 37, D489−D493. (44) Bodilis, J.; Nsigue-Meilo, S.; Besaury, L.; Quillet, L. Variable copy number, intra-genomic heterogeneities and lateral transfers of the 16S rRNA gene in Pseudomonas. Plos One 2012, 7, (4). (45) Kan, E.; Deshusses, M. A. Continuous operation of foamed emulsion bioreactors treating toluene vapors. Biotechnol. Bioeng. 2005, 92 (3), 364−371. (46) Bastos, F. S. C.; Rangel, A.; Castro, P. M. L.; Jorge, R. M. F. Biological treatment of a contaminated gaseous emission containing monochlorobenzene. Environ. Technol. 2003, 24 (12), 1537−1544. 6853

dx.doi.org/10.1021/es3049465 | Environ. Sci. Technol. 2013, 47, 6846−6854

Environmental Science & Technology

Article

(47) Fortner, J. D.; Zhang, C.; Spain, J. C.; Hughes, J. B. Soil column evaluation of factors controlling biodegradation of DNT in the vadose zone. Environ. Sci. Technol. 2003, 37, 3382−3391. (48) Hassan, A. A.; Sorial, G. A. Treatment of benzene and n-hexane mixtures in trickle-bed air biofilters. J. Air Waste Manage. Assoc. 2011, 61 (2), 201−210. (49) Mathur, A. K.; Majumder, C. B.; Singh, D.; Bala, S. Biodegradation of mono-chlorobenzene by using a trickle bed air biofilter (TBAB). J. Environ. Biol. 2010, 31 (4), 445−451. (50) Seignez, C.; Adler, N.; Thoeni, C.; Stettler, M.; Peringer, P.; Holliger, C. Steady-state and transient-state performance of a biotrickling filter treating chlorobenzene-containing waste gas. Appl. Microbiol. Biotechnol. 2004, 65 (1), 33−37. (51) Oh, Y. S.; Bartha, R. Design and performance of a trickling air biofilter for chlorobenzene and o-dichlorobenzene vapors. Appl. Environ. Microbiol. 1994, 60 (8), 2717−2722. (52) Cussler, E. L. Diffusion: Mass transfer in fluid systems; Cambridge University Press: Cambridge, 2003. (53) Mancini, S. A.; Lacrampe-Couloume, G.; Lollar, B. S. Source differentiation for benzene and chlorobenzene groundwater contamination: A field application of stable carbon and hydrogen isotope analyses. Environ. Forensics 2008, 9 (2−3), 177−186. (54) Fuller, M. E.; Mu, D. Y.; Scow, K. M. Biodegradation of trichloroethylene and toluene by indigenous microbial populations in vadose sediments. Microb. Ecol. 1995, 29 (3), 311−325. (55) Schroll, R.; Becher, H. H.; Dörfler, U.; Gayler, S.; Grundmann, S.; Hartmann, H. P.; Ruoss, J. Quantifying the effect of soil moisture on the aerobic microbial mineralization of selected pesticides in different soils. Environ. Sci. Technol. 2006, 40 (10), 3305−3312. (56) Baath, E.; Anderson, T. H. Comparison of soil fungal/bacterial ratios in a pH gradient using physiological and PLFA-based techniques. Soil Biol. Biochem. 2003, 35 (7), 955−963. (57) Pasteris, G.; Werner, D.; Kaufmann, K.; Hohener, P. Vapor phase transport and biodegradation of volatile fuel compounds in the unsaturated zone: a large scale lysimeter experiment. Environ. Sci. Technol. 2002, 36 (1), 30−39. (58) van der Meer, J. R.; van Neerven, A. R. W.; de Vries, E. J.; de Vos, W. M.; Zehnder, A. J. B. Cloning and characterization of plasmidencoded genes for the degradation of 1,2-dichloro-, 1,4-dichloro-, and 1,2,4-trichlorobenzene of Pseudomonas sp. strain P51. J. Bacteriol. 1991, 173, 6−15. (59) Werlen, C.; Kohler, H.-P. E.; van der Meer, J. R. The broad substrate chlorobenzene dioxygenase and cis-chlorobenzene dihydrodiol dehydrogenase of Pseudomonas sp. strain P51 are linked evolutionarily to the enzymes for benzene and toluene degradation. J. Biol. Chem. 1996, 271, 4009−4016. (60) Lee, C. Controls on organic carbon preservation: The use of stratified water bodies to compare intrinsic rates of decomposition in oxic and anoxic systems. Geochim. Cosmochim. Acta 1992, 56 (8), 3323−3335. (61) Lin, X.; Kennedy, D.; Fredrickson, J.; Bjornstad, B.; Konopka, A. Vertical stratification of subsurface microbial community composition across geological formations at the Hanford Site. Environ. Microbiol. 2012, 14 (2), 414−425. (62) Or, D.; Smets, B. F.; Wraith, J. M.; Dechesne, A.; Friedman, S. P. Physical constraints affecting bacterial habitats and activity in unsaturated porous media - a review. Adv. Water Resour. 2007, 30 (6− 7), 1505−1527. (63) United States Environmental Protection Agency. Johnson and Ettinger model for subsurface vapor intrusion into buildings. http:// www.epa.gov/oswer/riskassessment/airmodel/johnson_ettinger.htm (accessed 02/20/2013).

6854

dx.doi.org/10.1021/es3049465 | Environ. Sci. Technol. 2013, 47, 6846−6854